| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
B PathwayDepartment of Surgery, Downstate Medical Center, State University of New York, Brooklyn, NY 11203
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
B mechanisms. In this study, we aimed to determine protein domains that are essential for the immunologic function of PAP2 by mutational or chemical analysis. The protein activity for each mutant was determined by measuring TNF-
, IL-6, or IL-1 production in macrophages. Truncation of the first 25 residues on the N terminus of PAP2 did not affect protein activity whereas truncation of the last 30 residues of the C terminus of PAP2 completely inactivated the function of PAP2. Additionally, reduction of three disulfide bonds proved to be important for the activity of this protein. Further investigation revealed two invariant disulfide bonds were important for activity of PAP2 while the disulfide bond that is observed in long-form C-type lectin proteins was not essential for activity. Coupling the ability of PAP2 to up-regulate inflammatory cytokines via NF-B with its associated expression in acute pancreatitis, a condition with aberrant concentrations of inflammatory cytokines, we investigated whether PAP2 mutants mechanistically activate the NF-B-signaling pathway and demonstrate that preincubation with select rPAP2 mutant proteins affect translocation of this transcription factor into the nucleus. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
The C-type lectin superfamily is a large group of proteins which is characterized as having at least one carbohydrate recognition domain (CRD) (6). Over its course, studies have identified the CRD domain to contain highly conserved residues, motifs, and a consensus sequence (7, 8). More importantly, the implications of the CRD domain are broad and vary widely in function. This is exhibited by the classification of C-type lectins into 17 subfamilies, which is based on the protein’s overall domain architecture (1). Conceivably, the variability within the CRD domains produces a range of biologic functions, including ligand-binding sites for oligosaccharides and polypeptide ligands. Moreover, the structural basis pertaining to carbohydrate specificity has been well documented in many C-type lectin proteins (9). However, structural analysis of C-type lectin that are known to bind protein ligands has not been evaluated.
Many C-type lectins actually lack calcium- and carbohydrate-binding elements and thereby have been termed C-type lectin-like proteins. An increasing number of studies are beginning to show that "atypical" C-type lectin-like proteins are involved in regulatory processes pertaining to various aspects of the immune system. Examples of this include the NK cell inhibitory receptor Ly49A C-type lectin-like protein which is shown to complex with the MHC class I ligand (10), and the C-type lectin-like protein mast cell function-associated Ag which is involved in the inhibition of IgE-Fc
RI mediated degranulation of mast cell granules (11).
PAP proteins belong to the group 7 subfamily of C-type lectins, which are speculated to be involved in the regulation of the inflammatory process. Recent studies from our own and other laboratories suggest that PAP proteins are regulatory proteins that are involved in both the anti- and proinflammatory aspects of this process. We have previously demonstrated that PAP2 mediates the expression of inflammatory cytokines in macrophages through the NF-B pathway (see Ref. 12). We also showed that both antisense gene knockdown and Ab neutralization of PAP2 in rats with experimental acute pancreatitis caused a significant increase in disease severity (13). These findings corroborate other studies which showed protective roles served by PAP proteins during tissue injury (14, 15, 16). Taken together, these studies suggest that PAP proteins are key regulators of inflammation and their absence causes a dysregulated inflammatory process.
In this study, we investigated the importance of specific protein domains within the PAP2 protein by mutational analysis. An emphasis was placed on analyzing C-type lectin motifs, invariant cysteine residues, and a short N-terminal PAP domain which is speculated to be involved in the formation of fibril particles after trypsin modification (17). The biological significance of each of these domains was assessed by truncation analysis or site-directed mutagenesis. The activity of each mutant was determined in vitro by measuring the expression of cytokines from macrophages and comparing it to full-length PAP2. Because we previously showed that PAP2 activates the NF-B-signaling pathway (see Ref. 12), here we similarly examined the affect of mutant PAP2 protein activity on this pathway.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
The rat macrophage cell line, NR8383, was obtained from the American Type Culture Collection. Cells were cultured in F-12K medium supplemented with 15% FCS at 5% CO2 and 37°C. Before experimentation, macrophages were plated and grown to confluence. Experimental conditions included culturing cells with rPAP2 for the specified time period followed by the analysis of culture medium for secreted inflammatory proteins. Unless stated, the dosage of 5 µg/ml for PAP2 was used in all experiments. This dosage was selected because this concentration corresponded to the lower half of the log phase of preliminary dose-response experiments performed on NR8383 cells. Disulfide bonds in PAP2 were reduced with 1 mM DTT for 1 h before administering to NR8383 macrophages. The activities of mutant proteins were compared with full-length PAP2 and were represented as percent active.
Site-directed mutagenesis
Full-length PAP2 was used as a template for site-directed mutagenesis experiments. Mutagenesis was performed using the Quickchange II Site-Directed Mutagenesis kit (Stratagene). Mutant primers were purchased from Integrated DNA Technologies (Table I). PCRs were performed using a MJ Research thermocycler. Cysteine to serine point mutations were generated for cysteines involved in the formation of the three disulfide bonds: the long-form disulfide bond (Cys14–Cys25), the lower half bond (Cys42–Cys145), and the loop stabilizing bond (Cys120–Cys137). Cysteines involved in the formation of each disulfide bond were mutated individually. Mutant cysteine templates were subsequently used for another series of site-directed mutagenesis to generate cysteine mutants that affected more than one disulfide bond.
|
The 10 members of the PAP family were aligned at the amino acid sequence level using ClustalW software. Sequences were obtained from GenBank. A pairwise distance matrix was obtained by calculating the proportions of different amino acids. The matrix was then used to construct trees by the neighbor joining method.
NF-B nuclear translocation
NF-B nuclear translocation was quantified in NR8383 cells by visual fluorescent microscopy. Cells were plated on to cover slips to 70% confluence and cultured with 5 µg/ml PAP2His for 3 h. Cells were washed in PBS and fixed with 3.7% paraformaldehyde and permeabilized with 0.1% Triton X-100. Slides were blocked with 2% BSA for 30 min followed by incubation with 1/300 anti-NF-B for 1 h and Alexa Fluor 555-conjugated secondary Ab (Molecular Probes). Preliminary data from our laboratory has demonstrated that second messenger signaling can be evaluated at this early time period. Fluorescence was assessed by a confocal laser microscope.
Production of rPAP2 and mutants
Full-length histidine-tagged rPAP2 protein was purified following a similar protocol to the one described previously (see Ref. 12). Briefly, wild-type PAP2 or mutants were subcloned into the pET24a expression plasmid and transformed into bacteria. An overnight culture of transformed BL21-DE3 cells was diluted with 500 ml of Terrific Broth containing 100 µg/ml kanamycin and grown at 37°C to an OD600 of 2.0 and induced with 0.1 mM isopropyl-D-thiogalactoside for 3 h. The cells were pelleted, resuspended in resuspension buffer, and sonicated on ice. Bacterial lysate was repelleted and resuspended in wash buffer followed by solubilization buffer containing 6 M urea. Because the recombinant proteins were insoluble, they were purified using nickel-charged beads under denaturing conditions in resuspension buffer. The purified proteins were renatured in two dialysis steps: the first in 0.8 M urea, 0.2 M arginine, 300 mM NaCl, 30 mM Tris-HCl (pH 7.0), 10% glycerol and then in 0.8 M urea, 300 mM NaCl, 20 mM Tris (pH 7.5) for a minimum of 15 h.
Isolation of primary macrophages
Following nembutal anesthesia, primary macrophages were isolated from the indicated organ systems. In all purifications, cell viability was >95% as determined by trypan blue staining. Macrophage cell morphology was examined by light microscopy. Macrophage function, as determined by NO production, was determined by LPS stimulation.
Peritoneal macrophages. Rat peritoneal macrophages were obtained by i.p. injection of 15 ml of cold Hank’s buffer 4 days after i.p. injection of 10 ml of 4% thioglycolate as previously described (18). Macrophages were then cultured in 12-well tissue-culture plates. After 1 h of incubation, nonadherent cells were removed by washing three times with PBS. Adherent cells, consisting of 95% macrophages, were supplemented with fresh F-12K medium and incubated at 37°C for 2 h before experimentation.
Alveolar macrophages. Rat alveolar macrophages were isolated from lung tissue by bronchoalveolar lavage as previously described (19). Lungs were lavaged three times via a tracheal cannula with 10 ml of cold HBSS. The lavage solution was centrifuged at 1500 rpm for 15 min and the cell pellet was resuspended in F-12K medium and plated in 12-well culture plates.
Monocytes.
Whole blood was obtained from rats and collected in EDTA-containing tubes. Blood was diluted 1/3 with PBS and placed on a Ficoll-Hypaque gradient at 800 x g for 15 min. The buffy coat was isolated and contaminating RBC were lysed with a hypotonic RBC lysis solution (ammonium chloride) for 10 min. Cells were centrifuged (1500 rpm for 15 min) and pellets were washed three times with PBS. Cells were resuspended in F-12K medium supplemented with 15% serum and allowed to adhere to 12-well tissue-culture plates for 1 h. Nonadherent cells were removed by washing three times with PBS. Adherent cells contained 
95% macrophages.
Quantitative real-time PCR
Total RNA was purified from NR8383 macrophages by TRIzol extraction. Real-time RT-PCR was performed using the TaqMan One-Step RT-PCR master mix kit (Applied Biosystems) and the Applied Biosystems 7500 Real-Time PCR system. A total of 100 ng of RNA was used for each real-time PCR. Amplification (40 cycles) was conducted in a 25-µl reaction, containing 2x PCR master mix (catalog no. 4309169; Applied Biosystems), RNA, enzyme, and primer and probe (catalog no. 4309169; Applied Biosystems). Primer and probes used to analyze for IL-1, IL-6, TNF-, and β-actin expression are summarized in Table II. Probes contained the reporter dye 6-FAM at the 5' end and Black Hole Quencher-1 at the 3_ end Gene expression was quantitated relative to β-actin; relative expression of the target gene was calculated as 2_ddCt, where dCt is the difference between the Ct for the gene of interest and the threshold cycle for β-actin. In each experiment, the value of the relative expression of the control sample (untreated) was given a value of 1 and the expression of other treatments was plotted relative to the control.
|
IL-1, IL-6, TNF-, and IL-10 were measured in tissue-culture medium by respective ELISA kits (R&D Systems) in accordance with the manufacturer’s recommendations. The ELISA for these cytokines were sensitive to 50 pg/ml of the respective recombinant cytokine. Unless stated, all cytokine assays were performed after treatment with 5 µg/ml rPAP2 for 24 h.
Statistical analysis
All data represent a minimum of three independent experiments and are expressed as the mean SE (±SEM). Statistical analysis was performed using a two-tailed Student t test or 
2 test. A p value of <0.05 was considered as statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
To examine the relationship between PAP proteins across species, the amino acid sequences encoding rat, mouse, canine, sheep, bovine, and human PAP proteins were analyzed by multiple alignments using clustalW analysis (Fig. 1A). Depending on the species and isoform assessed, conservation of the primary structure for these proteins varied between 47 and 91%. This variation in the primary amino acid structure presumably permits various functions for these proteins. This is exemplified by the many diverse immunological functions that have been reported for select PAP isoforms (3). To further explore the relationship among these proteins, a phylogenetic tree was constructed for PAP (Fig. 1B). It should be noted that currently, three PAP isoforms are identified in mice and rats, two PAP isoforms are identified in humans, and one PAP isoform has been identified in canine, bovine, and sheep. As shown in Fig. 1B, branching at the root suggests a common ancient origin. A large cluster was formed by mouse (Reg3), rat (PAP2), sheep, bovine, dog, and human PAP. Interestingly, PAP isoforms from rodents formed their own independent clusters, with PAP1 clustering with Reg3β and PAP3 clustering with Reg3
. The final analysis reveals the greatest similarities exist between sheep and bovine PAP, mReg3 and rPAP2, mReg3β and rPAP1, mReg3 and rPAP3, and the human Reg3/human PAP subcluster forming a cluster with canine PAP. This phylogenetic analysis shows that the PAP family is conserved across species but disparities exist as demonstrated by tighter clusters among the indicated PAP isoforms. Thus, these analyses reveal that several PAP isoforms are more closely related than others.
|
Primary sequence analysis of PAP proteins reveals there are two major domains that make up the entire protein. With the exception of a small N-terminal domain which will be referred to as the PAP domain, the vast majority of the protein consists of the C-type lectin domain. The PAP domain is an extremely hydrophilic region consisting of the first 13 aa of the protein (Fig. 1A, blue box). Included within this region is a highly conserved trypsin cleavage site located at positions Arg11–Ile12 (Fig. 1A, double line). Studies show that trypsin cleavage of certain PAP isoforms generates highly insoluble fibril structures that are prone to precipitate (17). The biologic significance of this site in PAP proteins is not understood. The C-type lectin domain within PAP stretches from residues 14–149. Classification of proteins as C-type lectins is based on key motifs and residues. Conserved C-type lectin residues found in PAP proteins are denoted with blue stars (Fig. 1A). Within the CRD domain of PAP proteins, several but not all C-type lectin motifs are conserved. Beginning with the N-terminal region of the CRD domain, residues 14–25 are less conserved motifs that are present only in long-form C-type lectin proteins (Fig. 1A, red box). These residues form a β-hairpin that is stabilized by two less conserved cysteines which form a disulfide bond. Invariant cysteine residues are located at positions 42, 120, 137, 145 (Fig. 1A, denoted by black dots). The calcium-binding domain, "EPN" motif (positions 88–90), and glutamic acid (position 108) is present in all PAP proteins except rat PAP2 and mouse RegIII. The "WND" motif, also involved in calcium binding is absent in all PAP proteins. The highly conserved "WIGL" motif (residues 77–80) is present in all except for the PAP2 isoform, where the highly conserved glycine is replaced with a tryptophan residue, generating a WIWL sequence. The significance of this observation is not understood. Located on the C terminus of PAP proteins, between residues 120–149 (Fig. 1A, denoted by yellow box) is the C-type consensus sequence, which is commonly used as a landmark for sequence analysis. The loop region, which is the variable part of all C-type lectins, is depicted by black boxes in Fig. 1A. Assessment of the secondary architecture for PAP proteins using PELE analysis demonstrates these proteins all share a common secondary backbone—β11β22β3β4β5β6β7—which is in accordance with the secondary structure of C-type lectins (Fig. 1A).
Analysis of PAP2 tertiary structure
The outline for mutational analysis of PAP2 is shown in Fig. 2A. Because of the strong primary sequence conservation within the PAP group of proteins, we used the homology recognition program PHYRE to construct a tertiary structure for rat PAP2 (Fig. 2B). PHYRE modeled the tertiary structure for PAP2 on the x-ray crystal structure of human PAP (the overall structure of PAP2 displays the prominent features that encompass the C-type lectin domain (CTLD) fold (1)). This includes the considerable conservation of hydrophobic residues involved in the formation of the hydrophobic core, two antiparallel β sheets, two helices which flank each side of the structure, and three disulfide bonds. The loop region located on the upper half of the structure that lacks secondary structures and exhibits variability in amino acid sequence. With respect to certain CTLD proteins, the loop region is commonly associated with calcium-dependent carbohydrate binding and protein binding (20, 21). As shown in Fig. 2D, the PAP domain composes a small fraction of the entire complex and its location on the periphery of the overall complex demonstrates its independence from the C-type lectin fold. Immediately following the PAP domain is the less conserved C-type lectin domain which is only found in long-form C-type lectins (Fig. 2A, and Fig. 1A, black boxed sequences). This region forms a small loop that contains two cysteines which form the least conserved disulfide bond. The C-type lectin consensus region which corresponds to the last 29 residues is represented by the gray region in Fig. 2. This region encompasses a significant part of the highly conserved C-type lectin fold segment of the tertiary structure. Additionally, as represented in Fig. 2B, the consensus sequence encodes for β sheets (β6, β7, β8) which are important for the formation of the two antiparallel β sheets. The entire upper antiparallel β sheet is formed by β6 and β7, and β8 pairs up with β1 to form the second antiparallel β sheet. Additionally, the consensus sequence contains three of the four invariant cysteines found in all C-type lectin proteins.
|
We previously reported that rat PAP2 is important for macrophage activity by specifically up-regulating the expression of IL-1, IL-6, and TNF- (see Ref. 12). Due to its strong expression in acute pancreatitis, we speculate that this isoform is an important modulator of the inflammatory process. To study the functional significance of regions of PAP2, we constructed truncated proteins and compared its activity with wild-type protein. Truncation studies were conducted on the PAP domain, the long-form C-type lectin region and the C-type lectin consensus sequence Fig. 2. Truncation of the N-terminal domain of PAP2 (
N PAP2) containing the PAP domain and the C-type long-form domain was found to be superfluous for the transcription and secretion of IL-1, IL-6, and TNF-. As shown in Fig. 3, the activity of wild-type PAP2 and N PAP2 on macrophage cytokine production and secretion was comparable. In contrast, truncation studies on the C-terminal consensus (C PAP2) sequence completely abrogated the activity of PAP2 as demonstrated by basal levels of cytokine production by macrophages (Fig. 3). Unlike components of the N terminus, the C-terminal region of PAP2 is absolutely necessary for the function of PAP2.
|
There are three intrachain disulfide bonds in PAP2 that are situated in different locations within its structure and the primary amino acid sequence is shown in Fig. 4A. Currently, the importance of these disulfide bonds is not known. Two of the three bonds correspond to the invariant disulfides present in all CTLDs (Fig. 1A, denoted by dots above cysteine residues). The less conserved disulfide bond, Cys14–Cys25, is located within the long-form C-type lectin domain. The two invariant disulfide bonds, Cys42–Cys145, are positioned on the lower half of the fold, whereas Cys120–Cys137 is present on the upper half of the fold (Fig. 4B). In the truncation studies, we indirectly assessed the importance of their disulfide bonds because cysteines involved in their formation were within the deleted sequences. Removal of the disulfide bond (Cys120–Cys137) that presumably stabilizes the upper region of the fold (loop region) resulted in an 30–40% reduction in activity (Fig. 4C). Similarly, mutation of Cys42 or Cys145, thereby eliminating the bond located on the lower half of the fold, resulted in a 30–40% reduction in activity (Fig. 4C). Interestingly, the effects of double mutation studies removing both Cys120–Cys137 and Cys42–Cys145disulfide bonds resulted in a 67% drop in activity (Fig. 4C). In Fig. 5, full-length PAP2 was subjected to the full reduction of its disulfide bonds (1 mm DTT at pH 8, 30° for 30 min) to provide sulfhydryl groups. Comparing the activity of reduced PAP2 with unmodified PAP2 on the secretion of inflammatory cytokines in macrophages revealed a significantly attenuated response in the reduced isoform. This result further indicates that disulfide bond formation is essential for the function of PAP2.
|
|
We have previously demonstrated that PAP2 induces proinflammatory cytokine expression in both clonal macrophages (NR8383) and cells derived from primary sources, peritoneal, alveolar, splenic, and monocytes (see Ref. 12). We therefore investigated the effects of PAP-mutated proteins on macrophages obtained from primary sources. As shown in Fig. 6, full-length PAP2 induced TNF- expression in primary macrophages. In contrast, C PAP2 completely obviated TNF- expression. Similar to the results obtained with the clonal cell line (NR8383), point mutations of Cys14–Cys25, Cys120–Cys137, and the double mutant Cys14–Cys25 Cys120–Cys137 decreased TNF- expression by 30–40% for the single disulfide bond mutants and 60–85% for the double disulfide bond mutant in all primary macrophages.
|
B pathway
Previous studies from our group have demonstrated that the NF-B inhibitor Bay11 completely inhibited PAP2 mediated up-regulation of inflammatory cytokines in macrophages (see Ref. 12). Building on this, we investigated the ability of mutated PAP2 to affect NF-B-mediated translocation to the nucleus. As shown in Fig. 7, cells cultured with wild-type PAP demonstrate nuclear translocation whereas addition of NF-B inhibitor Bay11 completely inhibited PAP2-mediated nuclear translocation. In contrast to wild-type, PAP cells cultured with C PAP displayed a similar inhibition to that of NF-B inhibition demonstrating the importance of the C terminus with respect to PAP activity. Interestingly, cells cultured with C137 demonstrated a reduction in nuclear translocation compared with wild-type PAP. Similar results were obtained when cells were cultured with C147 and greater reduction in nuclear translocation was observed with dual mutated (C137, C147) PAP (data not shown).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
helices, two antiparallel β sheets, six cysteines, and a variable long loop region (1). We previously reported the C-type lectin-like protein PAP2 to be a modulator of macrophage function (see Ref. 12). In this study, we characterize the functional significance of structural elements present in PAP2.
The domain architecture for PAP proteins is unlike any other C-type lectin protein which is why they were accordingly situated into their own group, referring to group 7 of the C-type classification system (6). Interestingly, PAP proteins are the smallest protein reported among the C-type lectin family. Additionally, PAP proteins contain an extended CRD domain that is only associated with long-form C-type lectins (1). The long-form domain comprises residues 14–25 and encodes for a short β-hairpin that is stabilized by a disulfide formed by Cys14 and Cys25. Moreover, PAP proteins contain highly conserved residues that are essential for establishing the C-type lectin fold and a partial calcium-binding motif, though a complete absence is noted in rat PAP2 and mouse Reg3 (Fig. 1A). Furthermore, PAP proteins lack residues involved in carbohydrate recognition but contain four invariant cysteines that are important in the formation of three disulfide bonds. In PAP proteins, the domain corresponding to the loop region consists of several discontinuous loop structures that are separated by β sheets.
In this study, we assessed specific C-type lectin domains and residues by mutational analysis and analyzed their capacity to support the expression and secretion of IL-1, IL-1β, IL-6, and TNF- from macrophages. We initially constructed mutants that lacked residues 1–25 on its N terminus or residues 120–149 on its C terminus. To observe the significance of these domains on the proteins tertiary structure, we used a computer-generated model of PAP2 which was created by PHYRE. This program modeled the structure of PAP2 based on the crystal structure of human PAP. In our findings, we observed that the N-terminal deletion did not affect the activity of PAP2 whereas the C-terminal deletion significantly decreased the activity of PAP2.
We initially constructed a mutant that lacked the small N-terminal PAP domain which is the only region that is not related to the C-type lectin domain. When compared with full-length PAP, the activity of this mutant was comparable, thus retaining the ability to induce cytokine expression in macrophages (data not shown). To further investigate this region of the protein, truncated studies were performed on the ensuing C-type lectin long-form domain. This domain contains two less conserved cysteines which are involved in the formation of the disulfide bond located in that region of the protein structure. Interestingly, we observed that the combined deletion of the PAP domain and the long-form domain (residues 1–25) did not affect cytokine induction in macrophages. Thus, we speculate that this part of the protein is not involved in the putative macrophage-binding domain.
We also constructed a C-terminal mutant that lacked the C-type lectin consensus sequence (residues 120–149). Deletion of this region results in the removal of β sheets (β6, β7, β8). Analysis of the location and significance of these secondary structures within the proteins’ tertiary structure suggests an important domain which may explain the observed response in macrophages. Thus, we speculate that the loss of these critical secondary structural elements likely results in a gross structural instability in the tertiary structure of PAP2. Additionally, the truncation of this domain eliminates three invariant cysteine residues that are involved in the formation of two disulfide bonds: one stabilizes the upper fold and the other stabilizes the lower fold. Thus, because our truncation studies indirectly assessed the role of these disulfide bonds on the activity of PAP2, we next decided to take a more direct approach.
As previously mentioned, there are three intrachain disulfide bonds in PAP2: one is located in the long-form domain and two correspond to invariant bonds associated with all C-type lectin proteins. We assessed the importance of these disulfide bonds in the absence of any deleted structures and observed that reduction of the cysteines with low concentrations of DTT resulted in an absolutely nonfunctional PAP2. Additionally, because the long-form disulfide bond is located within the deleted N-terminal domain which did not have any inhibitory effect on protein activity, we speculate this bond is not important for PAP2 activity. However, due to the arrangement and internal location of the invariant cysteines, we speculate the resulting disulfide bonds associated with these specific cysteines are most important for the overall structure and function of PAP2. To test this hypothesis, site-directed mutagenesis of each individual cysteine was performed individually and sequentially. By generating a single point mutation that replaces a cysteine for a serine in rat PAP2, we have demonstrated that the loss of Cys42, Cys120, Cys137, and Cys145 markedly impairs cytokine production in macrophages. Absence of the Cys42–Cys145 bond on the lower pole of the structure resulted in a 30–40% reduction in activity and the deletion of the Cys120–Cys137 bond on the upper half of the structure similarly resulted in a 30–40% reduction in activity. Moreover, double deletion of both invariant disulfide bonds resulted in a 70–80% drop in activity. These findings were observed in clonal and primary macrophages derived from blood, lung, and peritoneum. In contrast, point mutations of Cys14 and Cys25 did not alter the function of PAP2 (data not shown). This further edifies our hypothesis that the absence of these critical disulfide bonds most likely affects the ability of the protein to properly fold. Additionally, because PAP proteins are secreted, we rationalize that the highly stable disulfide bonds would make these proteins capable of withstanding harsh extracellular environment. Perhaps, the deletion of these disulfide bonds makes the protein more susceptible to harsh external factors including shifts in ionic strength, temperature, and pH.
Within select C-type lectin proteins, the loop region has been reported to be important for protein function including: carbohydrate binding, protein binding, and homodimerization (22, 23). With reference to the C-type lectin expressed on NK cells, the loop region of the LY49 receptor interacts and binds to its ligand, the MHC class I receptor expressed on other cells (10). The loop region has also been shown to be important for the dimerization of C-type lectin snake venoms (24). For this reason, we were interested in analyzing the functional role of the loop domain in PAP2. Interestingly, this region coincides with segments displaying higher variability in amino acid sequence among members of the PAP family. The PAP2 loop domain is divided into two independent loop regions that are separated by β-sheets β4 and β5. The first loop, which corresponds to residues 81–92, appears to be the less variable loop whereas increased variability can be appreciated in the second loop domain (residues 107–119). Interestingly, within the second loop, rat PAP2 and mouse Reg3 share 92% sequence identity. Because the loop region is a highly exposed area, our approach to studying this part of the protein included a classic Ab inhibition approach: polyclonal Abs were raised against residues 81–119. Intriguingly, we observed that the incubation of full-length PAP2 with surplus polyclonal anti-PAP2 Ab did not attenuate the cytokine response in macrophages (data not shown). This suggests that the loop region is not pertinent to PAP2-mediated induction of macrophage cytokines. However, it is plausible that the full inhibition of this region by Abs does not fully occur. Potential problems with the Ab-neutralization experiment include the way the Abs were designed. Because a partial protein sequence was expressed and used in the immunization process, we speculate it may be possible that the partial peptide did not fold according to the loop domain in full-length PAP2, thus giving rise to Abs that recognize a linear peptide sequence or irregular epitope. Thus, the purified Abs may not recognize the tertiary epitope we originally set out to create.
We previously showed that PAP2-mediated up-regulation of inflammatory cytokines in macrophages was blocked by an inhibitor of the NF-B pathway (see Ref. 12). In this report, we demonstrate that PAP2 mediates the expression of the previously indicated cytokines through the NF-B pathway. The translocation of p65 into the nucleus of macrophages after culturing with PAP2 is revealed by immunofluorescence. After a 2-h treatment with PAP2, over 75% of macrophages stain positively for nuclear NF-B. Similar activation of the NF-B pathway at these early time points has been reported (25). In contrast, treatment with C-truncated PAP reduced NF- B nuclear translocation in a manner similar to that of NF-B inhibition. Furthermore, PAP point mutation at the cysteine residues either individually or together also demonstrate a reduction of NF-B nuclear translocation compared with wild-type PAP. These data demonstrate the importance of the C terminus and the disulfide bonds involved in PAP-mediated immunomodulatory function. The signaling of PAP proteins through the NF-B pathway has been previously described for other PAP isoforms. PAP1 has been shown to be a negative regulator of this pathway in a rat model of inflammatory bowel disease (16). Similar results were observed for PAP1 in a macrophages, and acinar cells (14, 25). An in vivo rat study demonstrated that a high dose of PAP1 activates the NF-B in hepatocytes (26). PAP activation of NF-B was also found to be important for the survival of damaged motor neurons (27). Thus, our findings corroborate previous reports and are novel in the sense that we demonstrate activation of this pathway by PAP2 and that structural integrity of the disulfide bonds is vital for PAP2 activity.
In summary, this study on PAP2 provides the first structurally based analysis of a member of the PAP protein family. Through protein truncation studies, chemical analysis, and point mutation studies, we demonstrate that the preservation of the CTLD fold is absolutely necessary for the function of PAP2 on macrophages. Moreover, deciphering potential biologic domains on PAP2 could have significant implications in designing therapies for diseases such as acute pancreatitis.
|
Disclosures |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
|
Footnotes |
|---|
1 Address correspondence and reprint requests to Dr. Martin H. Bluth at the current address: Wayne State School of Medicine, 8203 Scott Hall, 540 East Canfield Avenue, Detroit, MI 48201. E-mail address: mbluth{at}med.wayne.edu 
2 Abbreviations used in this paper: PAP, pancreatitis-associated protein; CRD, carbohydrate recognition domain; CTLD, C-type lectin domain.
Received for publication February 27, 2008. Accepted for publication May 16, 2008.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
B activation through a JAK/STAT-mediated mechanism in epithelial cells. J. Immunol. 176: 3774-3779. expression in hepatocytes. J. Pathol. 199: 398-408. [Medline]Related articles in The JI:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
). Three of the four invariant cysteines are found within the consensus sequence. B, The C-terminal truncated mutant. Note that deletion of the consensus sequence eliminates the last 30 residues, and two invariant disulfide bonds. C, The N-terminal truncate mutant. Deletion of the first 25 residues removes the PAP domain and the long-form C-type lectin domain. D, The predicted tertiary ribbon structure of PAP2 was generated by the structural analysis program PHYRE. The PAP domain and the long-form domain are located on the lower half of the structure. The consensus sequence is highlighted in gray and the encoded secondary structures are labeled β6, β7, β8. Disulfide bonds are highlighted purple, with the invariant bonds depicted by blue stars. The loop region is located on the upper half of the structure.
Ser substitution for Cys42, Cys120, Cys137, Cys145. Mutation of each disulfide bond were analyzed individually and concurrently. B, Tertiary structure of PAP2 showing the location of the respected disulfide bonds in yellow. The bond formed by Cys42–Cys145 (red) stabilizes the coming together of the N and C terminus and the bond formed by Cys120–Cys137 (blue) stabilizes the upper loop domain. C, The activity of recombinant mutant PAP2 proteins was compared with full-length PAP2. NR8383 cells were cultured with 5 µg/ml of the indicated protein for 24 h followed by the analysis of TNF in culture medium. A significant decrease in activity was observed for all three mutants, p < 0.05. Mutational analysis of the cysteines involved in the formation of the long-form disulfide bond did not affect protein activity (data not shown).
